Add community project to ditch Amazon covers

Why not add a community project to remove Amazon covers from LT? I don't think it's unreasonable that Amazon insists click on their cover takes your to their site, but it's a pain.

But if we created copies of Amazon covers - even getting them manually from Amazon if need be, that would mean we wouldn't have to do so.

While I'm at it, I do think much more could be done to encourage better cover use. Stats/Awards for uploading would certainly help, but more important would be a page where we could see all our books without covers, together with a quick link to all possible covers for those books - all on one page.

a page where we could see all our books without covers,

You already have this bit; if you go to your statistics page (link on your home page), click on "Book covers" on the left side. It will take you to a page which gives a breakdown of the sources of all your covers. Click on one of the categories to go to that subgroup in your catalog; from there you can go to the book page for each item to see the cover choices.

I think this would be an important step in moving away from Amazon data, something Tim has expressed concern about before.

And it seems pretty easy, too....

I've been working on it sort of on my own - I have all my Amazon covers in a collection and I'm slowly scanning or finding other sources for covers.

I think this is a great project - especially as it moves LT toward independence.

I'm also slowly replacing my old Amazon-sourced book data with library data when I can.

Same here, Helcura. I have a tag "Amazon cover" that I've been slowly weeding through and uploading covers for.

It would be wonderful to turn it into a project, though, get more people to join in and raise awareness about the need for non-Amazon covers.

Good suggestions. I looked at my book covers stats, I had 721 Amazon covers, but it's down to 622 now. It'll go down a lot more when I really focus on cleaning it up, but I only went through the first couple pages. I also like the idea to reward people for uploading their own covers.

whoa - I haven't been using amazon as a source for months, but it looks like I've been adding a bunch of books with amazon covers. How does that work?

7: If Amazon has a cover that matches your book's ISBN, it will show as an Amazon cover. You can pick from the member-uploaded covers, if any, or find a cover elsewhere (scanner, camera, on the web, etc).

awww, I thought I was virtuously pulling covers from libraries all this time. :( just going to have to make "change cover" part of my add books regime.

maybe this isn't such a great idea to keep as a default anymore?

I'm all for paring down the Amazon data, but the problem with member-uploaded covers is that many times they are just not good quality images. Call me shallow, but I'd rather have a good-quality Amazon cover than a blurry "homegrown" cover in my catalog.

I'm not trying to denigrate member efforts, especially since I'm not uploading any covers myself. What I really love is to have a scanner here at home, so I could scan my own covers and upload them. That's not an option, regretfully.

But there are many high quality user covers as well (Probably more than the low quality ones, in my experience). It's worth the time to see if you can find one.

>11 jjwilson61: Yes, I've made an effort lately to look for user-contributed covers for any book that I add (in addition to re-setting my default search to Library of Congress instead of Amazon). In fact, I just did a batch earlier this week, which is why the problem of low-quality covers was at the top of my mind.

It may just be a function of the particular books I was adding, which were all older editions that had pretty sketchy quality covers uploaded. No doubt for more recent/more popular books, it won't be as much of a problem.

And of course, it's wonderful that so many take the time to scan covers and upload them for all of us to use. Even to the extent of scanning covers of books you don't have in your catalog, Collectorator! That's dedication!

>14 Collectorator: Wait a minute! You mean you can use the Amazon picture URL and that's not the same as using the picture from Amazon? I'm not doubting you in the slightest, but that just seems ... weird.

By the way, I agree with your earlier message that picture scanners/uploaders should get a shiny badge!

I would love to finish getting rid of the Amazon covers in my catalogue. I prefer to use my default cover to having an Amazon cover, but when I change the Amazon cover to the default cover, it still shows up in the Amazon covers in my statistics.

Sorting through the long list for the few mistakes and old entries not yet changed is a pain.

Well, I'm doing my bit! I always scan the cover of any book I add if there is no good member uploaded cover already and (though I say it myself) mine are good quality. I'm currently going through my catalogue to replace any Amazon covers that I have in use.

>14 Collectorator:-16
I'm not sure you're allowed to do that... are we?

>19 Felagund:: The Amazon Terms of Service forbids Tim from automatically copying the Amazon image to the LibraryThing server. But remember that book covers are not copyrightable, and we're not bound to the same TOS as Tim is.

I would like to see a system whereby we could make Member-uploaded covers as duplicates of Amazon ones, preferably with a quality ranking, too. That would enable Amazon covers to be replaced in other people's libraries by member-uploaded ones, something that doesn't happen at the moment. It may be best if the person who uploaded the image is not allowed to make the judgement - perhaps better still if two or three people's votes are needed to make the switch.

if you go to your statistics page (link on your home page), click on "Book covers" on the left side. It will take you to a page which gives a breakdown of the sources of all your covers. Click on one of the categories to go to that subgroup in your catalog; from there you can go to the book page for each item to see the cover choices.

Thanks for this. I knew there was a search for no covers, but hadn't realised there was one for Amazon covers. But it would still be nice to be able to select an alternative cover from that screen in one click, rather than the three clicks using that method.

>20 markbarnes:

The image of a book cover may be available under fair use guidelines, but does that fair use rationale extend to cover Amazon's digital scan of the cover?

> 20

But remember that book covers are not copyrightable. . .

Where did you get that idea?

Furthermore, Amazon's TOS state quite clearly that images on their site are copyrighted, and that you cannot download any portion of the site without express written consent.

While I doubt that Amazon would go after a single, isolated violation, a concerted effort to by LT members to do an end run around that might well draw their attention and their ire.

>20 markbarnes:/23.

The covers ARE copyrighted, just not by Amazon.

>General.

I'm all in favor of this, though, I'd need to lock myself in my library with a scanner to be helpful. Maybe one of these days, I'll click that "Amazon-provided covers" link in my Stats page, and start scanning.

I normally only scan a book cover if my cover is not there, or the one that is there is pretty grainy.

Thanks for this thread :) When I started with LT, I didn't know there was any "harm" in selecting an Amazon cover so my library had almost 160 of those until I read somewhere that this was a no-no... Since then I've been meaning to go through those covers and change them where-ever possible but... well... good intentions and so on....

Now I've ditched more than 10 of amazon covers and will lose the rest within this week **decision made!!**

Slightly off topic: I wonder is there any other way to change the information source but to delete the book file and fetch it again from somewhere else? Deleting about 260 bookfiles (my amazon-sourced books) just to update them again is not a nice prospect...

Choosing an Amazon cover and using Amazon as a source are two separate things. Even if you use a library source, LT will still get the cover from Amazon if it can find one that matches the ISBN (since libraries do not provide covers). To use a user-uploaded cover you need to explicitly choose one after you've added the book from whatever source.

And the decision to use a library source or Amazon depends on what quality of cataloging data you want. Amazon won't provide Dewey Decimal Number, for instance, and will often mess up the author or put extraneous junk in the title.

Edited to Add: Oh, and to answer the question you actually asked, no there's no shortcut to re-add Amazon books from a library source. Although it would be cool to have a function that would "refresh" a book from a different source, and it has been requested before, it hasn't been implemented.

This may be relevant to 14-16, 19-20, 22-24.

>26 jjwilson61:: Thanks for the info :) Good to know there's still lots and lots and lots (etc.) work ahead with my library data **ironic sigh**
Wish I'd known all this about "the Curse of Amazon" before tackling my books but noooooooo, I had to input hundreds of books in a frenzy before condescending to read the discussion site........ ;)

@23: But remember that book covers are not copyrightable. . .Where did you get that idea?

Sorry, that was badly expressed. My point was the book covers are not copyrightable by Amazon, and in addition precedent and fair use would indicate that we can use them on LT for the purposes to which they are currently used without violating copyright.

Talvitar -

You're not alone. I think a lot of did the same thing when we started out. It's just so exciting to be able to catalog books so easily!

I actually like some of the extraneous stuff Amazon adds to titles (like what number and series) when I'm purchasing books, but it annoys the heck out of me in my catalog, which was what originally prompted me to switch to library sources. I still add a few from Amazon that I can't find in library sources. It will always be a valid resource, I suppose, just not my preferred one.

My preference is to use library sources, but certain kinds of books just don't get into academic libraries very often, so Amazon is the back-up - but I always clean up the record after adding.

The two Finnish library sources are very, very good indeed, even with books-in-English. It is almost always possible to find the record there -- for me the "problem" was that I began with my English novels and as the amazon was the default, I didn't even bother checking with the national or capital area library sources...

Live and learn :D

The real culprit here isn't so much that covers don't exist or aren't uploaded (although that is the case for some books, of course), but that the default is to use the amazon cover.

Your average user just isn't going to care that much about changing the cover - especially when it looks like it matches the book holding in their hands. It's an extra step that doesn't really accomplish much for the user (although it accomplishes a great deal for the site).

I'm maybe a mediumish power-user - I combine on occasion, I added used book stores, I read talk. And I never, ever bother to change the default cover unless it looks wrong. If LT wants to free itself from Amazon covers, uploading new covers is one step - but the bigger one is either using LT data to match the book entered with a member uploaded cover most of the time, or adding a step to 'add books' where you're asked to select a cover, with the member ones front and center. Until then, all the uploading in the world is likely to have only a minimal effect.

I hope Aerrin99 doesn't mind if I do my "minimal" best to help. Today I subtracted over 400 Amazon covers (replaced with member-uploaded covers) from my library. And I haven't started scanning yet. Funny. It doesn't feel useless.

There is a relatively new setup where it chooses a member-uploaded cover, 'best guess' by ISBN. But I don't know if that's default over Amazon, only if Amazon fails...?

I'm also working on getting high-quality, member-uploaded covers on all my books. I'm using a combination of tags and collections to help - all the books that I haven't checked and approved the cover are in a collection called Working On. If I've checked and found that there isn't a good member-uploaded cover available, it's marked with a tag - depending on whether I have the book and can scan it, _scan cover, _need cover, _make cover (that's for ebooks, which I want to have looking ebooky. Still figuring out what that will look like), _compare cover (there are member covers available, but I'm not sure if any of them match my book. Go find the book and compare the actual cover to what's available - if there's no good correct one available, scan it).

I also have the tag _amazoncover, which I applied by going to Stats/Book Covers/Amazon covers and powerediting it on. It makes it easier to find those, and combine dealing with them through the other tags (you can't search for or exclude a tag while you're in the covers view). However, I've got a heck of a lot of books to deal with - there's more than 3000 in Working On, 1400 don't have a _* cover tag (I haven't looked at them and decided what needs doing yet), and I still have some 600 _amazoncover tags (though only some 350 that haven't been tagged with a cover tag yet). Progress progresses, but this is definitely a long-term job.

If LT wants to free itself from Amazon covers, uploading new covers is one step - but the bigger one is either using LT data to match the book entered with a member uploaded cover most of the time, or adding a step to 'add books' where you're asked to select a cover, with the member ones front and center. Until then, all the uploading in the world is likely to have only a minimal effect.

This is absolutely true. What I'm proposing will largely solve the problem by:

• Allowing users to tag member-uploaded images as duplicates of Amazon images, which will only only allow them to be chosen when you add books, but could replace Amazon covers on existing books.


• Would give 'helpers' tools and incentives to upload and tag these images, to speed the whole process up.

My current stats are:

• Amazon covers: 1815


• Member uploaded, chosen by me: 1415


• Member uploaded, chosen by LT: 241


• No cover: 592

The large number of Amazon covers are largely due from my initial import, when I didn't know any better!

Am I going to get through 1,815 Amazon covers? Perhaps. But more importantly, if other users could help by tagging user-uploads as identical to Amazon images, then together that 1,815 could disappear to near-zero.

"adding a step to 'add books' where you're asked to select a cover, with the member ones front and center"

I really don't want another step in Add Books. If they could automatically use a member cover instead of an Amazon one, great. If we could change the cover from the Quick Edit, great. But mandatory extra steps in a basic process are bad.

I wouldn't mind the step. Even if the cover is correct, I still like to see if there are better choices (closer to mine in condition, etc.).

Ok, but some don't want extra steps every time. I get that too.

Maybe if a default setting could be set up for member uploaded covers preferred, or something, instead of making it another step.

> 40

But mandatory extra steps in a basic process are bad.

I agree. The fewer steps, the better. What would be really nice would be if I could go to the edit page, and add edits and change the cover without having to go back to that page.

#42 - You have my vote

And mine.

42> As it is, if you make changes on the edit page you then can _either_ click Save _or_ go to the change covers page - if you change your cover you lose your edits. If you click edit then immediately covers, the refresh from covers leaves you on the main page rather than back in Add Books (not that I want it to always put you there). Or whatever - covers is complicated from Add Books. So when I'm adding, I make it a separate step - add all the books and do what changes I can through Quick Edit (very few), then go to my catalog (sorted by date entered down) and change titles, authors, series, and covers (and whatever else is needed and editable through the catalog - most).

I go crazy when I get popped back to the Add Books page when what I really want to do is change the cover. What I do now is first click through to the book itself via the linked title (didn't used to be able to do that) instead of the edit pencil. From that page, I go to the edit page. Then Save will pop me back to the main page of the work. From there, I go to Change Cover, then back to Add Books via the top tab. I'd rather not have all the extra clicks, but the people who don't care about that and prefer to add book after book as quickly as possible would probably complain if it didn't pop back to Add Books.

But more importantly, if other users could help by tagging user-uploads as identical to Amazon images, then together that 1,815 could disappear to near-zero.

I'm a bit dubious about that idea, largely because I suspect I have stricter standards of "identical" than a lot of people -- a different publisher logo on the front, or different cover blurbs (as will often happen between the hardcover and paperback, for instance) will be enough to make me want to use an Amazon cover rather than a user-uploaded one, but I'm sure many people disagree, and would consider anything with the same general cover art to be "identical". I'd far rather use user-uploaded covers, and like most people am in the process of slowly going through my catalog and changing images as I can, but they have to be the right cover, not just a "pretty close" cover.

-47 Amazon is not that specific either - they will often change the cover associated with an ISBN as the relative popularity of different books with that ISBN changes. If you want that level of exactness, you are generally going to need to change the automatic cover anyway. That said, I don't think it is usually a good idea to change anything that is currently in a library. It would be nice to use a member uploaded cover as a default when adding books, though.

I actually think it is a great idea to be able to label member uploaded covers with an ISBN (when available). It is could be used to swap out Amazon covers for member upload covers when books are added. It would also be nice when searching member uploaded covers, as some works have a really cumbersome number of covers.

48>

Amazon is not that specific either - they will often change the cover associated with an ISBN as the relative popularity of different books with that ISBN changes. If you want that level of exactness, you are generally going to need to change the automatic cover anyway.

Er, yes. What did I say that made you think I wasn't aware of that extremely obvious and well-known fact? I'll happily use a blank cover if the correct one isn't available. But I don't want people willy-nilly deciding that two random covers are identical because they're "close enough" for them, and then swapping covers in my own personal catalog as a result.

I think the point that jlelliott was making was that the "automatic" swapping would only occur when someone was using the default Amazon cover. If they are, then they shouldn't be surprised to see it change. If you have carefully chosen your covers, then it wouldn't affect you.

50>

You are wrong.

I am in the process of carefully choosing my covers. Sometimes, the cover that I carefully choose is the Amazon cover. I do this when it is the only correct cover. I do this fully aware of the fact that it is subject to change in the future, but since I check my covers periodically, when this happens I will see it, and switch to a blank cover (or to the correct user-uploaded cover if one has been uploaded in the interim.) This proposal would swap out Amazon covers that I have specifically chosen to continue to use for user-uploaded ones that may not actually represent the same cover.

Look, I'm all for getting rid of as many Amazon covers as possible. But not if it means changing my data without me taking specific action to change it.

I do this fully aware of the fact that it is subject to change in the future, but since I check my covers periodically, when this happens I will see it, and switch to a blank cover

So what's the problem with the proposal? You're fully aware that you've chosen covers that are subject to change. They change. You check and switch to a blank cover if necessary.

Amazon has been imposing stricter restrictions on how their data can be used. There could easily come a day when Amazon covers are no longer an option, and the choice would be user-uploaded or blank. The transition will be a lot less painful if you have to deal with a few misplaced cover blurbs now--following the exact same procedure that you've already acknowledged is necessary--than if 90% of users wake up one morning to find that half of their covers have disappeared entirely.

-49, 51 I don't understand - if you are perfectly willing to switch a cover if Amazon has changed the default away from your preferred cover, why are you not perfectly happy to change it if LT does the same? This is especially confusing as the change coming from LT would happen once (assuming this happens as LT does away entirely with Amazon covers), as opposed to the frequent changes that occur with Amazon covers.

But productive of some good suggestions.

I wish I had the resources to throw at this. Believe me, it's important, even if we don't ditch Amazon. I want to do a mobile version, for starters. And they utterly prohibit their data being used on native mobile apps.

It seems to me that in the past year or two Amazon has gotten increasingly limiting and restrictive about that kind of stuff. Or anything to do with their data, really.

It would be easier/better to have an actual LT tool behind this movement, but if we rally enough I'm sure us users could do a pretty good job on our own. Spread the word, talk about it like we're doing here, inform users who might not know the limits of Amazon data, etc.

(and yes, I'm awake at 4:47 am. I haven't been to sleep. It's my dog's fault.)

Welcome, Tim! Very glad to hear a positive response. I guess what we're saying here is that we do have the time/resources when you're able to build the tool.

I wish I had the resources to throw at this.

So, have you thought about getting more resources? Other websites sometimes have donation drives, and they seem to do very well.

58: I think he means staff more than money?

>59 infiniteletters: Yes, but where do you think staff come from?

>60 _Zoe_: The staff stork brings them, silly!

Ahem. More seriously, I think some sort of tool would be quite welcome. After reading this thread over the past few days, I took the initiative to switch out as many of my Amazon covers as I could. Sadly, that was only about 30 of the 150+, mostly due to there not being an equivalent member-uploaded cover. But the process itself was incredibly tedious. I can't see many people being so dedicated to the concept of an Amazon-free LT that they would willingly undergo that. Having a tool to simplify it somehow would meet with much more success, I think.

The staff stork brings them, silly!

So, is it time to get Liam and Kate and Jasper doing some programming yet?

I've been trying to avoid Amazon covers for a long time now. I may be wrong, but I believe that my Amazon cover could be changed if Amazon switches to a different or newer cover for the work. Now there is even more reason to avoid them.

I've been trying to ditch the Amazon covers I do have, and switch to user covers or my own scans. That project led me to post a (pony?) request back in August at http://www.librarything.com/topic/72071. There was no response from TPTB, who were probably in the middle of a major feature addition. Perhaps, as it ties into this thread, someone would take a look? I don't care about the badges that others requested, but if having them got more people to help ditch the Amazonian covers....

>62 _Zoe_: If they're tall enough to see over the keyboard, why not?

Or maybe some Loxodonta africana staffers could be hired. I hear they work for peanuts.

When I first joined this site, I added almost all my (over 5,000) books with Amazon as the source, because I did not know better. I uploaded a lot of covers, mostly when Amazon's did not match my particular copy. I have been paring down my Amazon covers mostly because of the fact that they are subject to change. It's a relatively simple process, but I wish I'd just scanned them all when I had them in my hand because finding them on the (loosely organized--by genre and size) shelves is kind of a pain.

Is there a (hopefully similarly simple) way to adjust the data/source so it's not Amazon?

I wish my problem was choosing between Amazon and other sources for book covers. I have 2370 books showing no covers (out of 5700). With a non-working scanner, and the thankless effort of finding a copy of one of my books to make sure a member-loaded cover is the same cover as the one I have is the reason. For one of my Dune books, I had to wait 18 month after adding the book before a member loaded a cover that resembled mine. I have Amazon as a source for 1840 books.

Is there a way to see which books in my library have no member uploaded covers? I'm willing to start scanning covers, but would rather focus on a list of books that don't already have member-uploaded choices. I know I can see this if I look at each individual record, but I'm hoping to be able to generate a list so I can pull books and scan in batches.

>64 rosalita:: I've heard good things about outsourcing to Elbownia.

>67 ForeignCircus:: Unfortunately, there's no way to see this. It would really help if we had two pages: (1) A list of books with no member-uploaded images, (2) A list of books that HAVE member-uploaded images, but where we're using no cover or an Amazon cover. In this latter case, I'd like all the member-uploaded images to be displayed so I can choose the correct cover with one click.

A list of books without member-uploaded images would definitely help. I'd be willing to spend some time scanning covers, but pre-checking each title would really cut into my productivity.

The problem with a list of member uploaded books is it would not necessarily show "your" cover. Most of mine don't.
That said, I first tagged all 372 Amazon books. Now I am going down the list and changing all the covers. I scan and upload if the book is at hand, then remove the tag. If not, I MAY use a member uploaded cover or I may use the one used on other books without covers. The tags stay so I can check later when I return to Ohio and scan the lot. 320 to go.

It'd be nice if there were an easier, faster way to change cover selections. As it is now, I hit 'statistics' and then 'book covers' and then 'Amazon-provided cover' to get my list - okay so far.

But then in order to change my cover, I have to open each work page, then from /that/ click 'covers', and then finally chose my cover. So even once I have my list, it's three more clicks, and two more (rather slow on my connection) page loads to change every cover.

Is there a short-cut somewhere? Some way to streamline this so I don't have to load so many pages on my iffy internet?

You could put all the Amazon cover books in a collection, and in the display choose to show small covers. Click on the cover, select change cover. That's two clicks eliminated... right?

If you click on the cover picture from the catalog, it will give you a few options, including change cover. That'll get rid of one click.

Last weekend I started the SLOOOOOOOOOW process of scanning in my Amazon-only book covers. (I must say: for people who are totally AR like me about having the RIGHT book cover, scanning is the best way to go.)

I expect to be done by the next US Presidential election. :)

75:

I never knew about that shortcut. Thanks.

ETA: And 73, and 74. (Perhaps I wouldn't get it if I wasn't told several times?)

> 73, 74, 75 - THANK YOU! You folks are wonderful - I didn't know that option existed, and I shall make use of it.

(Now if only I could do so in a light box, without leaving my catalog. What? I thought a talk post wasn't complete without a pony request. ;)

I, too, had forgotten about the ability to single click for this menu!

So, I called up my Amazon covers list from the stats page and am going through to see if any of my 581 Amazon covers have member uploaded options...I notice that I can "right-click/open in new tab" on the change cover link from the exposed menu so I don't have to "back" to the catalogue and wait for it to load.

(Fun playing with this but I haven't found any covers to replace yet - guess I will be getting out the digital camera again this weekend.)

I cataloged about a thousand books before I got my scanner working, but they are now boxed up in anticipation of the Big Cross-Country Move (which is always impending but never seems to have an actual date attached to it). However, if other people start scanning, maybe I can match things up.

We just got another dozen boxes of books out of the storeroom, so as those are cataloged, I will scan away.

If only LT could pull the books out of storage and repack them for me. (Why stop at a pony? I want a Percheron!)

80: If only LT could pull the books out of storage and repack them for me. (Why stop at a pony? I want a Percheron!)

Exactly.

That may actually be a Pegasus! But I want one.

Oh, me too ...
(heartfelt sigh ... )

I have scanned most of my books that did not have loadable covers. I guess we should all scan covers of our books as local shareware, if we are to truly be free from Amazon. I dread the day they will ask us to pay for the privilege of using their covers as default. But then again, if they were to do the same thing for old and rare books of ours, it could also raise some funding for the site.

Boo, hiss. I'm discovering some of my books have cover damage (coffee stains, etc.), which is why I used the Amazon image instead. I shall have to rely on the kindness of strangers (who are tidier than I).

Oh, I totally clean up all of my cover scans. Creases, dings, scratches, stickers. Whatever I can photoshop away (or Paint Shop, in my case) I do.

Ditto. Irfanview is actually pretty good at cleaning things up. It also makes it easy to make it lighter - and since the cover image (what is displayed for the book) is noticeably darker than the thumbnail (for some weird reason), I try to make the thumbnail very bright.

I go to the Gimp for complex stuff, like actually adding a title to some of my hardbacks that only have it on the spine. Yeah, not accurate to my book, but a heck of a lot more useful than a blank sheet of color...

(both of those are free programs, and very capable)

See, I'm one of those who leaves every dent and scratch completely visible. In fact, I'll often upload my own cover even if there already is a member-uploaded cover, if my cover is worn/faded/ in some way different looking.

I like to go through my catalogue and actually see *my* covers, creases and stickers and all. It makes it that much more personal. It would feel weird to me if I read a book that had a noticeable crease on the cover, and then looked at my cover on here and it looks all perfect. That'd be weird.

Ah. While I feel that the cover on LT should be an idealized, 'new' version of the cover, while my book has had individual events change it. Even for my own books, leaving aside the question of being useful to other people for cataloging. That does explain some of the covers I've seen, though, that people chose to put up their individual versions. Thanks!

>88 jjmcgaffey: ...adding a title to some of my hardbacks that only have it on the spine.

If the name is on the spine and I can get a decent pic I upload the spine image as the cover.

Re: cleaning up the covers - I've been known to clean up the actual cover before taking its LT cover pic (removing stickers, taking goo-gone to the tacky places) and I will fiddle with the contrast etc. but more extensive photoshopping seems like a lot of work.

The problem with a spine image is that LT sizes purely by width. Which means that a spine image, unless severely cropped (which would work, but look odd) will be normal width and a mile high. What I generally (try to) do is get a spine image, cut out the title (and author if present) and paste it onto an image of the front. I haven't actually uploaded any like that yet - still working on them.

What I do is scan the cover, rub out (with the clone tool or just a paintbrush, depending on the background) any egregious markings (and yes, I remove stickers and as much residue as possible before scanning), boost contrast and gamma, resize and upload. It takes between 2 and 5 minutes per book, depending on how marked up it is. Seems worth it to me - YMMV.

The covers without any title I'm still working out how they'll look - so far I like two covers with just a title superimposed, and one with the spine data superimposed. But I'm still working on standardizing my workflow on those.

>93 jjmcgaffey: The problem with a spine image is that LT sizes purely by width. Which means that a spine image, unless severely cropped (which would work, but look odd) will be normal width and a mile high.

I also do an "angled" view if I want to be able to see the color of the cover as well - examples here:
http://www.librarything.com/catalog.php?view=PortiaLong&tag=TEMP+-+angled&am...
- this doesn't seem to create the same GIANT covers that the spine-only pics do:
http://www.librarything.com/catalog.php?view=PortiaLong&tag=TEMP-GIANT+COVER...

I clean up my cover before scanning, removing all stickers, plastic and all.

And I scan each and every book because I want *my* covers, even if the book is uncreased and new.

I have even had a situation where I bought a book at Amazon, but the cover on Amazon was wrong (With Lost City of Z where the Amazon image had the old explorer on the cover, and mine has David Grann, or the other way around). Better to scan all my own!

If there is a high resolution member-uploaded copy of my cover, I will use that. Otherwise I scan everything "as is". I only make adjustments for brightness and contrast. I inherited a lot of books from my mother, and she identified all her books by writing in Braille on the cover. Those books mean a lot to me, so I want to see their covers.

For books that are read but not owned, I look for the cover I remember from when I read the book. If there isn't a good member cover available, I search online for one. I get the best results from AbeBooks. That is an ongoing project for me.

I was fortunate enough to have learned about Amazon covers after entering only a few hundred books, so it wasn't too hard to go back and change them. I deal with covers as I enter books, so now all of my covers are member-uploaded.

For spine images, I open the book flat on the scanner bed and then crop out the back cover leaving the spine and the front cover. Here's an example from yesterday: http://www.librarything.com/work/206492/covers/53920063

I like having "my" covers simply because it will make it faster for me to find it later. I'm very visual; I'll search for the color of the spine before the title because that's how I remember the book first.

I clean up my scans partly as a service to LT, as I figure no one's going to want to use my scan if it shows damage. (I know I wouldn't.) I've spent up to an hour touching up a cover; particularly tricky are covers with metallic or foil lettering or backgrounds, or comic book covers that were mishandled when I was younger.

If LT has a high-quality cover that matches my own edition exactly, I'll use that, but I still like to scan my own for my personal use.

>98 saltmanz: That's dedication! On behalf of all of us non-scanners, thank you for your hard work. I will only add that as a user (and not to date a contirbutor) of member-uploaded covers, I am not deterred by images that show wear or tear, or stickers.

The only covers I won't use are the ones that are so excessively blurry or dark that the image quality is significantly inferior to the dreaded Amazon cover. Unfortunately, it seems a fair number of my books have only these lower-quality scans. But as more people decide to scan, I have high hopes that there will be better alternatives sooner or later, and I intend to keep checking.

94> Yeah, the angled covers have enough width to not shoot way up.

94, 97> Both of those would work - maybe I'll use one or both methods instead of doing the title-onto-blank-cover business. It would be more similar to the book. Still thinking about it - nothing's quite clicked yet.

96> My problem is frequently that several covers are familiar! I've got a whole bunch of books tagged _compare cover, which means that there are several high-quality scans available that look familiar and I need to look at my physical copy of the book and see which one I have. And several times I've found that my copy has yet another cover, not yet scanned.

In order to facilitate the transition from Amazon to user-supplied cover pictures, it might be interesting to add an item to the Book Covers statistics: "Amazon-provided cover" for which a "Member-uploaded cover, best-guess for ISBN" exists. We'd need a simpler name for it, too :-)

I'd be careful about using the amazon cover images directly, as in my experience they tend to have a lot of whitespace padding left and right of them. Also, they often have the "look inside" image superimposed

>102 Donogh:.

You can use the URL parameters to add/strip all that Amazon book cover glitz. There was an article about such in the Autumn 2008 issue of 2600: The Hacker Quarterly (I think that was the issue, as I remember it being in the last one I read).

When I snag a cover from Amazon (or elsewhere on the web), I don't upload it straight to LT, but run it through Photoshop first to crop out that extra space.

Spurred by this discussion, I spent the last several days ridding my catalog of all 841 Amazon covers it had attached to it. Next up is scanning covers for books that had no member uploaded option or that had none that matched my exact edition.

Brava, fleela! I'm working on it too, though much more slowly. A New Year's Resolution!

Also if you use the Go Inside feature there is a small thumbnail of the cover at the top left of the screen. If you click on it to save, you will get a full size .jpg with no 'Look Inside " banner on it.

Oh, thank you! That's a great trick! I knew it had to be there somewhere, because the covers never have it when LT brings them over, but didn't know how to access it.

>103 aethercowboy:

Wasn't there an LT blog post about that back in 2006 or 2007? I could swear that that's how I found out about hacking the Amazon image urls. Or, possibly, it was linked from LT somewhere way back then. I've always associated the post with LT, at least?

Anyway, here's an example for everyone of an easy way to copy the image without having to view more than the main search/book page:

is the tiny picture on the Amazon search/book page. In Windows, right click to copy the image location and get the URL: http://ecx.images-amazon.com/ images/I/41zzP4Y5xvL._SL110_PIsitb-sticker-arrow-sm,TopRight,8,-14_OU01_.jpg

You can paste that into the box and remove everything between the first stop after /images/I/ and the jpg:
http://ecx.images-amazon.com/ images/I/41zzP4Y5xvL._SL110_PIsitb-sticker-arrow-sm,TopRight,8,-14_OU01_.jpg

You end up with http:// ecx.images-amazon.com/images/I/41zzP4Y5xvL.jpg which is usually a medium quality, 500px tall version of the cover image.

Not sure if this really should be posted here like this, but I figure it's fairly general knowledge, and I personally prefer to get better quality versions than Amazon's, myself. But it's the quickest way I've found to get the cover.

(edited because of the dang link parser messing it up)

There's an even better way that can be done within LT, without ever going near Amazon's site. Take this example from Alexander the Great. All you do is right click on the Amazon image and choose "Copy Image Location". In Internet Explorer I think you have to choose "Properties" and the you can select the URL.

The URL will look like this: http ://images.amazon.com/images/P/0143035134.01._SY190_SCLZZZZZZZ_.jpg

All you need to do is remove a few characters after the .01. and before the _ so it now looks like this. This gives you the fullsize Amazon image:
http ://images.amazon.com/images/P/0143035134.01._SCLZZZZZZZ_.jpg

You can import that straight into Librarything's URL field. Here's a comparison of the three images first from Amazon's page, second the Amazon cover on the LT page, third the large version you've created (not all books have super large covers like this):

>110 markbarnes: You know, somehow it didn't even occur to me that the Amazon cover thumbnails would be linking back to Amazon, but it's so obvious that they do -.-

That's a nicely eliminated step, though, should the thumbnail cover be a good match for your book.

I'm still not comfortable with doing that for my own books, though I have done a few as a temporary measure until I can get a scanner. I just don't think that the Amazon versions resize very well.

Any books I own OR can borrow from the library, I scan the covers myself. I've gone through a LOT of the books I've catalogued by don't own, and grabbed covers from ebay, Abebooks, wherever.... I do have a few where I've grabbed the Amazon cover and uploaded it, just to get a "member uploaded" cover on there, but those are tagged specially so I know it still needs a better cover.

#110 nice. Apart from the "copyrighted material" header and footer, which I'm fairly sure doesn't appear on the actual book cover.

>113 reading_fox: The copyright notice occurs on a minority of Amazon covers. If you notice it would be there even if you chose the Amazon cover using the normal LT method.

112> I've even recently made a "Cover Scans" collection, where I can upload cover scans of books I'm giving away as gifts, or which I would otherwise have no reason to catalog.

I had already scanned all my books for my offline catalog (BookPedia on Mac) and now I have uploaded them all. For my wishlist I have searched online for replacement covers or used the user uploaded ones....

I am happy and proud to say, I am amazon-cover free!

Truly divine.
As we depart these wintry climes for Sunny South Florida I must quit scanning but will continue deleting Amazon covers, when and where I find them. Scanning will have to wait.

Will get to see a brand new snow blower imported from Canada for use in Global Warming. My son says the cab was shipped 26 days ago but has not arrived. Hope to get there in time to see the test.

back to covers.
Is there a way to upload a cover to be used as a member's default?

I am now "Amazon Cover" free. At least that is what Tim's search engine says.

http://www.librarything.com/profile/flabuckeye/stats/covers - yep :)

I've worked my way through my catalog - just have a few where there is no cover out there, or no good cover. I've got those tagged to scan. Mostly Amazon free ...

Bravo, both of you!

*Back to work for me!*

Oooo, I've been lazy... I've only ditched +50 covers so far, 108 yet to go...

Incidentally, do you know if there's any such stat where we could see the LT-total amount of Amazon-covers?

I offered a scanner to someone up-thread, but she can't use it. So: anyone out there want to scan, but can't due to the fact that you don't own a scanner? If you pay the shipping, you can have mine. It's an HP All-In-One. The printer no longer works, but the scanner's great. I do NOT have the original software disks for it. It's an HP PSC 2355. Here are the specs from HP.com:

Dimensions
17.32 x 14.3 x 8.2 in

Weight
12.2 lb

Supported operating systems
Windows 98; Windows 98 SE; Windows 2000; Windows Me; Windows XP Home; Windows XP Professional; Mac OS 9.1; Mac OS 9.2; Mac OS X v 10.1.5; Mac OS X v 10.2.3; Mac OS X v 10.3

Leave a comment on my profile if you are interested.

Just to do my bit, I scanned two or three dozen covers last night. Now for the other 260+.

I've been scanning some covers every day over the holiday period. I have to say most of them seem to be for books with low ownership numbers but it will help a few other people at least.

CD, those are actually the most helpful! Thank you, even if I don't have any of those books!

I've been neglecting my duty! I still have many books with Amazon covers, but most of them are ones I don't own any longer. Must hunt down online covers now.

I have a few of those too. In the meantime, I'll be making a dent in Amazon covers for books I still do own.

I created (yet another) Greasemonkey script, this time to help me spot which books have Amazon covers and therefore should be fixed. With it, I get something like this:



The script is here for those that are interested: LibraryThing Amazon Image Highlighter

What I'd like is if Tim could code this kind of functionality into the site (probably have to be opt-in). But I can imagine that these images come from a ton of different places in the code so that could be a challenge.

>129 brightcopy: Thank you, jpierce. I just installed your script. Since the original post, I've been working my way through my covers -- replacing the amazon covers with my own scans (mostly) and other member uploads. Your script gives me a very handy way to identify covers.

Yeah, you can see all your Amazon covers by going through your statistics page.

You can also use the search function in your library to access the Amazon covers (does the same as clicking the Amazon-provided link in Statistics) - use the search term coverstats::amazoncover.

>131 Collectorator:, 132, 133

I've always used the stats page to identify amazon covers, when I allocate time to changing covers. Now, I am motivated to scan the covers of books that are handy when I'm browsing my catalog, because the distinction is obvious. I'm very happy with jpierce's script so far.

An unintended bonus is that I can see when the smaller cover icon (used on LT's home page and profile page for recent additions and connection news, and in my catalog) is still using the amazon cover. It's an LT bug that's been around a long time, easily fixed by "change cover" and re-selecting your uploaded file -- but not always obvious, unless the default cover image is different from the scanned cover image.

In FIrefox it is possible to turn off Amazon images. I'm now working on my husband's computer, and having them there is driving me crazy,because I have to REMEMBER to change covers instead of having the big empty spot remind me.

134> That was pretty much my thought. I wanted something that would should "hey, change this!" while I was poking around my library.

There was one part of the script I was unsure about, though. I have it looking for "images.amazon.com". But I'm assuming sometimes it will link to other amazons like "images.amazon.uk", right? Anyone got an example of a book that might wind up doing that so I can test out better coverage?

>136 brightcopy:.

You could probably just do a RegExp for "amazon" in the image URL. I doubt there would be a large instance (if any) of the LT image servers having the word "amazon" appear in it.

Or, if you want to be safe:

var regexp = /\.amazon\./i;

http://www.librarything.com/pics/fav-amazon.gif

;)

Yeah, I could get more complicated to avoid that, but that's just why I'm a little hesitant to go with the shotgun.

135> Thanks, MarthaJeanne. I hadn't thought of doing that -- mostly, I guess, because I'm visual person who enjoys seeing the images (amazon or not, even though I hope to eliminate amazon images from my catalog).

>138 brightcopy:.

The last example I gave wouldn't match on that.

I'm not 100% on JavaScript RegExp, but if you wanted to be EXTREME, you'd have to write something like this:

var regexp = /^https?:\/\/({-\w\.}+)+\.amazon\.({-\w\.}+)+/i;

(replace {}'s with square brackets)

If I got that right, it should get anything with "amazon" in the domain name, including www.amazon.example.com. Such is life...

Edit: LT parsed out a lot of my RegExp.

Actual implementation left to reader. (,;;,)

140> Actually, it would have matched, as posted. But I'm guessing LT probably did its frustrating habit of stripping out the backslashes you put before the periods. ;)

It's not a problem to code it if I wanted to. I'm a pretty old hand at regexps. I just haven't yet mentally committed to "any site with amazon in the URL counts." I'll probably come around, though.

>141 brightcopy:.

My expertise is in Perl RegExp, so writing them in JavaScript requires some careful thinking to take into account everything it can't do.

Of course, my expertise for anything else is in Perl.

129 > This script is great. Thanks.

Amazon-provided cover (933)

Holy Cow! I need to get busy

I've brought mine down from over 2000 to about 350 in the last three months. It just takes working at it, a little at a time...

38
Am I going to get through 1,815 Amazon covers? Perhaps. But more importantly, if other users could help by tagging user-uploads as identical to Amazon images, then together that 1,815 could disappear to near-zero.

I think this touches on an importance piece of solving this problem. A community project to play the 'match game' with covers, linking Amazon covers and Member covers that are the same (let's say 95%+ the same, for sake of argument). We've seen from experience what the power of the LT community can do, all we need is a code framework around this.

It doesn't solve the whole issue, of course, but it helps to build information.

I have to agree- I am working on switching my covers from Amazon to member covers, but would love to be able to generate a list of books in my library for which the Amazon cover is the only option so that I can work on scanning those in to share...

Are you guys weeding out Amazon covers from your Wishlist and other collections where you don't actually own the book? Just curious.

I was surprised last night when I found myself down to just 35 Amazon covers remaining in my Your Library collection.

>148 saltmanz:

My wishlist contains only books that I actually plan on buying/getting. Once I got them, their covers change (when I do not forget) to remove the Amazon one.

I'm weeding them out of everything. I still have 1300 or so to go, but I'm getting there.

I'm trying to figure out how to get the covers to register as high quality and can't figure it out. It doesn't matter how I scan the cover, they always display on LT as fuzzy. Can someone point me at the specs I should be using in my scans?

151> It has nothing to do with any "specs", and everything to do with the size (width and height) of the image in pixels. LT doesn't (currently) display any covers at a height greater than 250 pixels, so that's a good minimum height to shoot for.

Thinking about it further, it could also be the JPG compression you're using; if you compress an image too much, it will turn muddy.

Can you post an example or two?

148> Yup, I am fully Amazon free. I usually don't wish for a particular edition, so I just use any user uploaded cover for wishlist or none at all. I did start with my library though...

>153 saltmanz: Thanks, size is a spec, at least to me.

I've been scanning at 600dpi and resizing down, with and without JPG compression. I think the problem is I should be resizing down to 250px height and making sure the text is clear at that size, rather than using a bigger size and letting LT resize on the fly.

http://www.librarything.com/work/9407183/55079802 is a work with only one cover, the one I uploaded. I know I can get the text to be sharper, I just have to figure out how.

I scan at 150dpi, which results in over 1000 pixels high for a mass market paperback, and I don't have an issue with LT resizing my covers.

Looking at your link, I wouldn't really call that blurry. It's not super-sharp, but it's okay. When I hear "blurry cover" I think of something like this: http://www.librarything.com/work/3804157/covers

LT is going to end up resizing your covers anyway, because of all the varying sizes used on the site. The 250 I quoted is (as far as I know) only used on the largest setting in Cover View. Images on a work/book page look to be 214, and images from the covers page are 190.

Ah. Thanks for all the display sizes. I hadn't thought to check the cover view display.

This info really needs to get out to newbies--I've been on since last July, and I'm only now starting to weed out my 800 covers...Aargh! I wish I'd known sooner...

>158 BeulahChurchLibrary:

Oh, it's all about that ladder of knowledge/familiarity/obsession that almost everyone seems to go through. It is not made easy by the foundational connections with Amazon data that's now potentially deep in the code. I have been a member since Oct 2005 and am only now realizing the extent of Amazon-sourced trouble. So now I have tagged books "scancover" for which I could not find a member-uploaded cover (a process which took a weekend :) and will use the opportunity to also verify info on the book that I may not know without looking at it. I am even thinking of creating some collections/tags around the concept of "Data verified" or something similar as I go through this process.

>156 saltmanz:

Thanks for the tips on scanning. I actually work with digitizing historical documents and it can very difficult to especially get text to be clear--my eyes go directly to even the most subtle compression artifacts now! It's a combination of all the different variables (pixels, dpi, compression) that one may need to test and practice before hitting good all-around like for paperbacks, large coffee-table books, etc. Something I will need to do once the scanner comes out of storage...

I still have about 900 Amazon covers left.

Slowly but surely.

148> Yes. Obviously I can't scan them, so I either pick a member-uploaded cover (any one that looks good, since I don't care about edition until I have the book), or if there aren't any (or aren't any of a quality I want), I'll do the right-click and save image location on an Amazon cover, save and upload that. I usually go in and pretty it up - sharpen and lighten - but not always.

159> When I started this, I created two collections - Working On and Cover Done. I put all my books into Working On and went through them gradually; ones that had a good-quality member-uploaded cover (that matches my book - there are some labeled _compare cover) go into Cover Done, ones that have a poor-quality member-uploaded cover get labeled _scan cover (or _make cover, for ebooks and the like where I need to do some manipulation), ones that have an amazon cover (and the coverstats::amazoncover plus Power Edit made this easy) are labeled _amazoncover and then I went through and checked whether they had available member-uploaded covers or I needed to scan them. Or if neither worked, the procedure above.

Right now about a third of my books are in Cover Done. I've been working preferentially on the _amazoncover books, to get those cleared up. Later I'll work on the others. Some of them are actually done, I just got them good covers before I started using Cover Done - but there's no way to tell except to look at each individual book. And that takes a while when you're close to 4000 books...

Now I try to take care of newly-entered books on entry, so they don't add to Working On. Though when the library booksale happens again in May, I suspect there'll be a bunch that I can't handle all at once. Ah well.

On scanning, I would always go for a larger size image than is currently needed. If you upload an image that is (for example) 1000 pixels, LT keeps the large image, and then generates smaller versions of it. If in the future it needs (for example) a 350px image, it can go back to the 1000-pixel image and make a 350-pixel version.

These original-size images are used already in widgets, which can be any size you want. If you look at this widget, created as a demo, and make your screen as big as you can, and I think you'll be impressed with the quality of the covers (all uploaded by me). The first cover is an Amazon one, which looks great elsewhere on LT but small here. The next one is a very high quality image I was fortunate enough to get from a publishers website, which is superb. The third is a scan of mine which also looks great at this high resolution.

I therefore scan at 200 or 150dpi depending on the size of the book, basically as high as possible that keeps within the 600k filesize.

> 163

The maximum upload size is now 2M.

Just another resource or two for covers. Quite a few can be found by searching the title in Google Images - sometimes great big covers that are really nice looking. Also, fantasticfiction.com has a surprising number of books and sometimes a great selection of covers (not always, but still ...). Finally, barnesandnoble.com and sometimes other booksellers will have covers. I've been able to find all but a handful of covers from places other than Amazon (I don't like that the Amazon covers are always a little fuzzier after you've uploaded them to LT). For me this is quicker than scanning my own books.

I am now using the tricks explained above to grab the best quality Amazon cover into the URL field, for now. However, either the scanning quality is not always great or close to 500 pixels is not cutting it because some still do not look wonderful. For instance, they never get labeled as "High Quality."

Hmm, should I keep grabbing decent covers or wait to scan my own at say 200-300 dpi (and maybe also correct any lingering bad data while holding the book in my hands)? Decisions, decisions...

189 Amazon covers left, about 35% of my library, when I first found this thread it was at least 300+

At 200 Amazon covers remaining, I am now officially below 40% from that source. I started keeping track of my progress between Christmas and New Year's, when I had 320 covers -- just over60% of my library -- from Amazon. (That said, my progress has been slow lately. A good friend was a bit delayed in gift-mailing by a particularly nasty case of the flu. But send it she did, and it proved to be an expansion pack for The Sims.)

166>

The program that labels covers as "High Quality" runs overnight (or something) so you'll never see it applied as you upload. Check back, though, and (if your cover was over 142 or whatever pixels tall) you'll see it.

166> As someone said above - some of the Amazon covers are supplied by individual merchants. Most of them have a big version (if you take out the SY190_ in the middle of the URL, you'll get the big one instead of the thumbnail - that's part of why I save and upload rather than grabbing), but some have only a thumbnail-sized image so they'll be rather fuzzy (since they're being upsized instead of down).

For myself - if I can't scan right now (like, if I don't know where the book is), I'll grab a decent cover and use that, but leave the book in Working On and tag it for scanning. That gets rid of the Amazon cover and puts a reasonably useful one into play, and I can make a good cover later when I get the chance. You could tag it 'upgrade cover' or something, if you don't feel the cover you grab is good enough - I should do that, actually, to indicate that the cover is correct but not high enough quality.

Right... I've lost about a 100 amazon-covers -- 62 to go. 57 are my own books so that problem is solved by scanning them. 1 is a wishlisted book so I don't mind that one so much -- when I get my hands on it I'll scan it. Only 4 books are older RNO-titles and finding the exact matches in the internet has so far been unsuccessful.
So 4 will remain -- or I'll try to find them in some library and scan the covers...

I've asked this earlier but nobody reacted so I'll try again: Do you know if there's any such stat where we could see the LT-total amount of Amazon-covers?

I wonder whether Tim could quickly rig up a script to speed up this progress. One of my publishers (IVP) has high quality covers of all their books, using a simple URL like this: http://www.ivpbooks.com/largecovers/9781844742592.jpg

It would be relatively easy for Tim to write a script that scanned LT for ISBNs that began 9781844 or 085151 (or whatever the IVP codes are), and then hit the IVP server for those images. The could perhaps be put on a page somewhere for manually checking by us. Users could then create similar data for other publishers by typing in URLs like http://publisher.com/covers/image.jpg?isbn=MAGICNUMBER and entering in the unique publisher part of the ISBN that matches the URL.

Yeah, but doing that without permission would be dicey. At some point someone would type in a library that uses Bowker or Ebsco covers, and we'll get sued.

Hmm, that's a shame. I just visited five publisher sites at random and they all used ISBNs to return images. Perhaps moderate submitted sites?

• http://www.christianfocus.com/images/items/9781845505486.jpg


• http://images.gnpcb.org/products/big/9781433514975.jpg


• http://www.bakerbooks.com/Media/PubComProductCatalog/9780801072444.jpg


• http://www.zondervan.com/media/images/product/large/0310326281.jpg


• http://www.hendrickson.com/images/oversize/9781598562293o.gif

I found two more "Amazon covers" this morning for books added this month. Both show my default cover on Library Thing. This happens about twice in a hundred books. Usually paperbacks.
Back to zero, again.

I recently discovered that if you apply your chosen default cover, the book stays on the Amazon cover list. If, however, you apply a different 'default' cover, it goes into Blank Covers Chosen by You. Better, I think.

Hmm, that's a shame. I just visited five publisher sites at random and they all used ISBNs to return images. Perhaps moderate submitted sites?

That's very interesting. I think the answer is to tie this to a "publisher account" feature we have planned. If we know:

1. The ISBNs that a publisher has
2. The URL of the publisher

And we have permission, we could really go to town.

Interesting, but a check now and then keeps the number at zero. Using Amazon only as a last resort helps.

FWIW
While entering a pile of books this morning I noticed that some Amazon titles had a plain black or really dark brown cover. It looked like my default cover at quick glance. So I changed my default cover and entered more. Now some of the newly added books have the Amazon plain cover but the cover count is still zero. Will change them anyway and scan when I get them to Ohio.

If we know: 1. The ISBNs that a publisher has; 2. The URL of the publisher; And we have permission, we could really go to town.

That would be terrific! (Though regarding permission I'd have all publishers in, and let them use robots.txt or meta tags to opt out. That's the established way to prevent another site from spidering yours. This isn't a copyright issue, of course, but a website terms of service issue.)

I am celebrating today, because I have less than 300 amazon covers in my catalog -- down from over 1500 when this thread started.

Here's a toast to markbarnes' original proposal!

Finding the motivation to tackle the last 300 is my next challenge.

I'm down to 155 books with Amazon covers. No idea how many of those 155 books are in storage though.

Bravo, both of you! I am working on it too, but have a long way to go.

I've gotten rid of a lot more, but it gets agravating to wade through pages of my default cover looking for ones that are really still Amazon.

Somehow some of my books went back to Amazon covers. This is so weird, because for one, it was automatic, secondly it were mostly Dutch books, and thirdly I had uploaded covers myself (own scans). It happened on 10 books, I have changed them back now...

183>

I wasn't totally clear on your message so you may already know this....apologies if if I'm telling you something you've already done.

You can view your cover breakdown via this link, just replace USERNAME with your user name (this is part of the zeitgeist in your profile)
http://www.librarything.com/profile/USERNAME/stats/covers

Then, if you click on Amazon-provided covers, it will show just the books in your library with Amazon-covers. Or direct link here (again replace USERNAME with your user name):
http://www.librarything.com/catalog/USERNAME&deepsearch=coverstats::amazonco...

No wading required.

185> except that if you apply your default default cover to a book, it still shows up in the Amazon list. So MarthaJeanne has replaced most of her Amazon covers, but they still show up when she goes to the links you mentioned.

If you apply a different default cover, it gets off the Amazon list and goes to Blank covers chosen by you - I've used that quite a few times to clear my Amazon list. I've got 323 Amazon covers left...still working. Mostly because I can't find the books to scan them! Gotta get to organizing my shelves...(goes off muttering).

#186
except that if you apply your default default cover to a book, it still shows up in the Amazon list

It's not working like that for me - I have 28 books listed as 'no cover' at the moment, which use my default, and none of them are showing up on the Amazon covers list. (This is apart from the 'blank covers chosen by you' where I have over 500).

>184 divinenanny: I've noticed that if I edit my book just after uploading a cover, the cover doesn't stay attached to my book. I have to re-select it.

>183 MarthaJeanne: MarthaJeanne, I installed brightcopy's script (in message 129), and it has helped me. I know that greasemonkey scripts slow the pages down, but it's been hardly noticeable and really worth it in my case.

p.s. I've thanked brightcopy earlier, I believe, but would like to again.

185 >

Thanks, I figured there was something I was missing.

the cover-stats that LibraryThing gives isn't completely correct, at least for me. I went through some of the list that it gave me, of Amazon covers I have, and some of them are actually covers I've uploaded... Guess it just didn't pick up on it? I don't know.

I'm down to 77 Amazon covers, NOT counting my wishlist. I don't like to mess with book-data (titles, covers, whatever) on wishlist books because I don't really feel like they are "mine" yet.... Or something.

I've been playing around with some covers, and I think I've found out why my experience in #187 is different from what jjmcgaffey is talking about.

I now have one default-cover book on my Amazon covers list, and one on my 'Member-uploaded cover, best-guess for ISBN' list - in each case that's the cover with which it was originally entered. So, as far as I can make out:

-Enter a book with no cover, and it will show up on the 'no cover' list.
-If it gets a 'best guess' cover when entered, and you replace it with the default, it will show up on the 'best guess' list.
-If it gets an Amazon cover when entered, and you replace it with the default, it will continue to show up on the Amazon list.
-If it gets an Amazon cover when entered, and you replace it with a non-default blank, then it shows up on the 'blank cover chosen by you' list.

Since the defaults showing up on the amazon list appear to have all been manually changed from Amazon covers, making sure to pick a non-default blank when changing would avoid the problem arising in future.

Thanks to the info I learned in this thread, today I've gone from 1,900 (25% of my works) having Amazon covers to half that. Thanks, everyone, for pointing out why to do this, and how!

It's still disappointing that covers on new books still seem to default to Amazon. I just added ISBN 1857926757 using Overcat, and it used an Amazon cover (for the wrong book!) in preference to a really good quality user-uploaded image.

>193 markbarnes: I agree and now changing the cover is a necessary part of adding a new book.

193>

And yet I was pleasantly surprised when a couple of my recent manual additions got attached to member-uploaded covers. (I wanted generic editions, so did a manual add in order to not have any ISBN/publishing data attached to the books.) It was really obviously not amazon covers, because of the quality difference, and one of the books' covers wasn't available from the amazon list.

Since I didn't use any ISBNs for those books, I was expecting it to default to the first Amazon cover in the list. :)

No, the Amazon covers are purely linked to ISBNs, so if you don't put them in - yay, it went to the member-uploaded! That's good, wish it would do it some more.

LibraryThing goes in a cascade of failure.

1. A cover you selected or uploaded
2. Amazon, by ISBN
3. Member covers from someone else, by ISBN

I don't think swapping 3 and 2 would actually be a good idea. On balance, I think it would reduce quality. But I may be wrong.

T

198>

In your library, the "member uploaded, best guess by ISBN" option accounts for 124 covers.

( http://www.librarything.com/profile/brightcopy/stats/covers )

199> Dangit, lorax. Why'd you have to reply. I deleted it and reposted to the better thread as quick as I could. ;)

I started with over 160 amazon covers. Now I only have 4 left -- those I don't own and cannot find decent images anywhere on the internet. Too bad. Maybe some day I'll try to get my hands on the books from a library, or something...

Anyway, no more amazon covers of the book I really own. Took me over a year. "Fast" is most definitely _not_ my middle name... :)

Congratulations! Welcome to the club :)

Just a suggestion, a lot of Author websites have cover images, thats where I get some from. I'm one of the ones who like perfect covers, if there isn't any member covers uploaded I'll search the internet first for good cover images, Publisher websites, Author sites, review sites, other book sites (Goodreads, IBDB, ISFDB ibookdb, Fantastic Fiction), then scan my own (and clean it up if needs be) if I really can't find the right one. TBH I don't do anything about the Amazon covers - I keep thinking I'll get round to it one day, but even then all you have to do is save the image to your computer then upload it here.

Argh! three had gotten past me! I just scanned the two that are still near the computer. The other one has a default cover for now.

I have too many Amazon covers and not enough time. But each time there's activity on this thread, I go and replace a few!

Slowly making my way through them for my own catalogue. "Just" another 400+ to go. Sigh ;-)

#204, 205, 206 - You've 'guilted' me into working on mine, again. LOL

#207 I finished changing mine a few weeks ago - a massive project but worth it!

I was happy when I got them under 1,000, and then I forgot about it. Now it's down to 920. I still have a long way to go.

Yay, clamairy, you can do it (ultimately). Hoot! Hoot!

(Where are those Pom-pons when you need them).

I'm laughing! I just got it down to 875, and now I have to take a break because I'm getting sloppy. (Sloppy, as in picking covers that are 'close enough.')

I'm working on getting covers for books that don't have them at all first, or have crappy ones, but when that's done (just 200+ to go!) the Amazon covers are next. Fortunately the crappy covers often *are* Amazon covers, so I'm chipping away at them in a small way.

Uh oh. Just when I was warming to cover replacement, I find that I'll have to put it on the back burner in favor of replacing old-style "other authors."

#213 - Luckily I only have 17 of those. *phew*

213> Yahoo! Finally! I have, literally (I think) thousands, so that's going to take a while. But I'll keep chipping away at covers, too.

Basically all my amazon covers are new books I've entered and haven't yet scanned - I still have a lot of unsatisfactory covers but a current total of (ow! bigger than when I last checked) 20 Amazon covers. Chip, chip...

ETA - only 306. Not nearly as many as I thought. Good, quickly done (for generous values of 'quick').

#215 - Two weeks!

Got my Amazon covers down to 2.
:-)

Wow! Way to go, MM! How long did it take?

Little at a time when the mood struck - probably a couple of months.

ETA: maybe more like three.

Bravo, MerryMary!

Don't forget - I'm retired. I have more "idle" time for dinking around than most of you.

Okay, I misread that as 'drinking around.' I thought, 'MerryMary, at this hour?' ;o)

I still have 850 to slog through. Sometimes I get lucky and some nice member has already found and uploaded or grabbed a decent cover. Mostly not, though. :o/

I've been wrangling 3 grandchildren for a week (Momma and Daddy come home tonight). I could probably use something stronger than iced tea!

I am getting there. Thank goodness I have a small library compared to many others at LT.

Huzzah! My own library is done. Yet another 100+ to go for wishlist and such, but obviously that makes less sense. I might delete from it, or get another edition, etc. Probably will do them anyway, in the end.

#225 - Congrats! I'm jealous. I ran out of steam last week, but I'll pick it up again this week at some point.

Oh, you guys, I'm jealous too! I still have 3,491 to do. I really wish I would have known about the badness with Amazon covers when I first entered the library and had them all in hand and scanable. Now, I have to go find them on the shelf, bring them down to the scanner, scan and reshelve. Still working on it, though, because I'm committed to this goshdurnit!

> 226/227: Don't feel bad, it's all good as long as we're chipping away at Amazon dependency.

One of the things I've found useful for trying to do this is the TinEye image search engine:

http://www.tineye.com/

If you're using Firefox of Chrome, there's even a plugin so you can just right click on the image and search the net to find snaggable copies of the cover. It doesn't work very well if the cover image isn't original to the book though.

I noticed that sometimes Amazon changes the cover of a book when I chose the correct cover in the past! I'm going to change all those into my own covers...

Yes, and picking an Amazon cover may also change your ISBN.

I know (there was a warning on the site :) ), so I've taken care that that didn't happen with any of my books.

I just changed 103. That's enough for today. Only 747 to go.

Only 502 Amazon covers left to change!

clamairy, good luck :D

Mine's done - at least, until I get around to entering this box of books on my floor. Then I'll have a few more, until I clean up my scans (I do them as part of entering - enter the book, check if it's got a good cover, if not scan the book and store the scan) and enter them. The cleaning up and uploading take more time than the scanning, so splitting the task works well for me.

BarkingMatt wrote:
Yes, and picking an Amazon cover may also change your ISBN.

Which isn't necessarily a bad thing - I was picking covers for some books in my library recently, and the two times picking an Amazon cover prompted the change ISBN popup it turned out I'd originally entered the book with an erroneous ISBN.

Anyway, I believe I'm missing something here. Why do people want to get rid of Amazon covers? The little Amazon links are very unobtrusive, aren't they? Is there a reason beyond dislike for the company?

Well, I can't speak for everyone, but... Amazon data is very unreliable. From what I've been able to understand, if something changes on the Amazon site, it also changes here.... So, for instance, if you pick a specific Amazon cover and then Amazon goes and changes that cover, it will change your cover with no warning. (That's what I've gathered, anyways, I may be wrong).

Also, overall it's simply a good idea to have as much information as possible coming from the actual LibraryThing website (like covers), instead of depending on an outside source that you have absolutely no control over.

It's not dislike per se, but rather too much bad data. And Amazon has a nasty habit of changing your cover when they change theirs. You will suddenly find your carefully chosen cover has disappeared, and some other (movie-in, perhaps) has miraculously appeared. Some people don't like that.

Thank you (pl). I'd very much dislike having one of my covers changed like that, too, but hasn't happened to me yet (that I've noticed), and I had presumed the images were static, perhaps grabbed to here.

Hm. Would it be very naughty to take cover images from Amazon and manually put them here?

I don't think so. As I see it Amazon can't claim any rights to those covers.

It's not so much a case of Amazon reaching out to LT and changing your cover, but that when you choose an Amazon cover in LT what is stored in your book record is a URL to a file on Amazon.com. So when they change their file you cover changes.

Re. #239: "Would it be very naughty to take cover images from Amazon and manually put them here?"

In a few cases where the cover was very hard to satisfactorily scan (like where it used a lot of foil) the Amazon image was better than what I could come up with, and I ended up using their image, while pixel-editing out the grey "copyrighted material" lines at the top and bottom.

 

I buy so many of my books from B&N that I have no issue with saving their lovely large covers (which can be found on most recent editions) to my hard drive, and then uploading them to LT. I did notice that if I just 'grab' the image using the B&N url that the quality is not as good. Not sure why that is.

I've still got 1,744 to go (but then I have already done 2,879!). One additional tip that might help some, is to use covers from Google Books when they're available. It's a bit complex, but there's some great quality images, and it's easier than scanning.

(1) First, make sure Google Books is listed in your Quick Links.
(2) Then, from the change cover page, you'll be able to see if the book is listed on Google. If it is listed as "Full View" or "Partial View", you might be able to get a high quality cover.
(3) Click on the Google link and see if the cover you need is displayed. If the cover you need is listed, click on it.
(4) Once you get the full cover displayed in all its glory, you need to find the image address, which Google make a little tricky. If you have Firebug, or the IE Developers Toolbar, you could use that tool. If not, continue below.
(5) View the source code for the page (right-click and choose 'View Source').
(6) Near the beginning of the source code is the image link. It's on the second full line. Once you know where to look, it's easy to spot. It looks like this: http://books.google.com/books?id=-POJuDejO_8C&pg=PP1&img=1&zoom=3&am.... Copy that link to the clipboard.
(7) Back in LibraryThing, paste that URL into the appropriate box on the change cover page, but delete the &w=??? from the end. Removing this code ensures your image displays at the native resolution.

So, it's a bit involved, but once you know what you're doing, it takes less than a minute. You can get some very high quality images, though they do have a 'copyrighted' watermark in the bottom-right corner:

>236 AndreasJ:
Andreas, yesterday I discovered that Amazon had changed the cover displayed for one of my books to the back of the book! This really won't do! I immediately picked a member-uploaded cover to replace it, but it's pretty poor quality, and I will be scanning a better copy as soon as I can.

244> Why can't you just right-click on the image and copy the link?

>246 jjwilson61:. Try it and see. It doesn't work - presumably Google deliberately have stopped it through overlaying the image with multiple DIVs.

It used to work, but hasn't for quite a while.

>244 markbarnes:

Although I'm a big Firebug fan, that's a little complex for many people (well, including me, to be honest). I find the easiest way to grab GB covers in Firefox is Tools | Page Info. Click on the Media button.

In the top half of the dialog box, you'll see the many things that have loaded on this page. I usually head straight to the bottom, and in the several I've looked at over time, the cover is the third or fourth from the bottom. Once you find it, click on the "Save As" button and then you're set.

Isn't the policy of Amazon towards sites and applications using their data also part of the reason for this project? I know of one application and LT, that they cannot make apps for mobile access to your collection because Amazon doesn't allow its data to be used on mobile devices (like that). If we lessen our reliability on Amazon data, it can only be a good thing.

Yes, though Amazon just changed their policy on mobile apps (so there's one on the way...when Tim et al have time). That was also the driver behind OverCat as a source.

This thread's getting awfully long. Shall we start a new one?

@#251... Good news! Right know I am keeping two databases, one on here, and one offline, using Bento. This is partly so I can have an offline copy on my iPhone/iPad, partly so I have a version of the data that I control. Good to know that the first 'problem' might get a (partial) solution soon :D

I've just blitzed a few hundred books which needed adding to LT, and am back on covers again. Here's my current method, which is giving me a rate of over 100 covers an hour. It should work for you if most of your books are relatively recent and still in print.

(1) Use the link from my Stats page to show only those books with Amazon covers.
(2) Sort this list by publisher (you may need to edit or change your view to show the publisher field)
(3) Switch to covers view. This will save quite a bit of time.
(4) Click on the first image. The pop-up window will appear very quickly, and tell you the publisher. Open a new tab and find the publishers website.
(5) If the publisher gives decent cover images, follow the steps below. If not, find the next publisher (temporaily switch to the grid view if you need to), and try again.
(6) CTRL+click on the edit covers button to open it in a new tab.
(7) Find the book using the publisher's search box. Open up the book's page.
(8) Find the best quality cover picture. Often clicking on the first cover shown yields a popup with a better cover. Then right-click and copy the image URL to the clipboard. You can sometimes save a click by right-clicking on an "Enlarge image" link and choosing copy link to clipboard.
(9) Go back to your "edit covers" tab, and paste in the image URL. Then close the tab, click on the next image, and repeat from step (6).

>249 legallypuzzled:
In the top half of the dialog box, you'll see the many things that have loaded on this page. I usually head straight to the bottom, and in the several I've looked at over time, the cover is the third or fourth from the bottom. Once you find it, click on the "Save As" button and then you're set.

I understand why you prefer to use Firefox's, Tools | Page Info | Media, but there's no need to save then upload the image (unless you need a local copy, of course). Instead, just right-click on the address line and choose copy. You can then paste the URL into LT, which should save several seconds.

255: I do that a lot, but I personally prefer to have a copy on my computer as well. I have full faith that LT will keep the covers I upload, but there have been glitches in the past, where covers that have been uploaded disappear or don't show up, and I like having back-ups just in case. (Like, for example, the so-so quality covers that got resized badly waaaayyy back when, I know that I can just re-upload them if I have a copy on my computer)

I need to go through my "Amazon cover" tag again. Most of those are books I no longer own and haven't been able to find decent covers online, but I haven't looked in awhile.

I've done a bit of searching on this theme and haven't found an answer. Has anyone decided what an ideal cover resolution is for LT? I often resize covers for my blog, and could easily make a LT version at a custom resolution.

Anyone know?

Has anyone decided what an ideal cover resolution is for LT?

The higher the better (within reason). Covers are used at very many sizes throughout LT. Although most of those are relatively small, using the blog widget it's possible to specify covers at very large sizes. Certainly I'd say 1000x1500 would not be excessive if you have that quality (though most of what I upload is about half that).

1000 x 1500 may be coming close to the 2 meg upper limit of what folks can upload though. edit: Would depend on the file format of course.

The max size that's presently displayed on LT is 250px high by about 175px wide. If you upload a bigger one, it's stored, and it's possible that at some later date we can see bigger covers; also, for most covers, higher resolution resizes to a better (clearer) image. There are some problems with really big images (in the 1000px+ range) - moire seems to show up sometimes. And yellow-on-red is always problematic (and unfortunately popular in covers of a certain age). So...yeah, no, there's no single resolution that's perfect for everything. But if you upload as big as you can, then look and decide if the resulting image(s) look good to you, that's about the best you can do. And LT has some tweaked resizing algorithms, I believe, so what works elsewhere may not work on LT and vice versa.

The max size that's presently displayed on LT is 250px high by about 175px wide.

Well, no, I don't think that's entirely right either. It also depends on resolution, number of colours, etc. I've ersonally upoaded pictures up to 1000 px high.

My cover image scan process includes a scan at 72 dpi (screen resolution) 100% size saved as a PNG. My HP scanner software is not great but does allow for cropping and rotation before saving. The result has been satisfactory for me. I don't like using a lossy compression (JPG) when additional processing (thumbnail) is going to be performed on the file. The end result is indeed a JPG but making a smaller JPG from a JPG can introduce display artifacts which make the image look worse than it might otherwise.

James

261> I said _displayed_. That's the largest image on the Covers page. You can certainly see the difference between a cover uploaded at 250px and at 500px - I generally upload at 600px high. But that's what's fed into the resizing algorithm to produce the 250px image that LT displays.

Google now have an image search facility that works from a URL. If there's an Amazon image that you'd like to use, but it has a copyright watermark, you can copy and paste the image URL into Google and search. Hopefully that will help you to find a watermark free version of the image on another site. To use it, go to google image search and click the little camera icon that now appears in the search bar.

Looking at my earlier message, it appears that since December 2009, I've added 2,244 images (but still have 1,160 to go!.) Hopefully the new search method will speed things up a little.

I had just over half my covers as Amazon covers (about 1,500). I'm down to 2! I'll try and dig up the books this weekend and get those cleared out.

#262 - My HP scanner software is not great but does allow for cropping and rotation

If you are a Windows user, try looking into Paint.NET. It's freeware that has been invaluable to me on my quest (I tend to have to darken my scanned images and add some contrast).

Another vote for Paint.net ... it's a great program, and free!

The URL to download it, however, is www.getpaint.net rather than just "paint.net" (which appears to be registered to some paint manufacturer) ... which is somewhat confusing.

Oh, and one caveat ... you need to have at least XP with SP3 installed to run it ... most folks won't have a problem with this, but my younger daughter's system is an older e-Machines with a motherboard that, for some reason, can't handle SP3 (it goes into endless reboot cycles - real ugly), so we've never been able to install that on her machine.

 

#265,

Thanks for the suggestion but I've been a Mac user since 1985. The "HP Scan" software is decidedly substandard but I'll probably have to stick with it.

For Mac, you can try Gimp, or user Preview for cropping and rotation (I use my Canon software + Preview).

Every time someone adds to this thread it motivates me to knock off a few more amazon covers from the pile.

Re. #267: "a Mac user since 1985"

Gee ... and I keep hearing about how the Macs are perfect, infallible, amazing, incredible, etc., etc., blah, blah, blah, for dealing with graphics ... how confusing that you're having problems with processing covers!

 

#269 Ditto - only a few at a time, but it all helps!

270 > A Mac is great for graphics because it was originally designed for graphics display, not for word/number display. HOWEVER, that and $1000 will get you a nice edition of Illustrator ...

I was going to add something snarky along the lines of #270, but didn't wanna start a PC vs. Mac (vs. Linux) war.

Now go knock out a few more covers!

I didn't say that I was having a hard time. HP Scan is not a great program. HP does not make the hardware and software with the same level of quality it once did, at least for the Mac market. The problem I was referring to was keeping natural colors with any settings offered in the HP Scan program. The preview scan looks good but the colors get washed out the moment the scan completes.

I also have Photoshop and use it often but sometimes it is overkill for the task. This HP scanner doesn't have a TWAIN driver for the Mac so I can't scan within Photoshop as I might with another scanner.

James

Down to one (and 8 missing covers) and that's on a work I need to look at as the series is mixed together:

http://www.librarything.com/catalog/theapparatus&deepsearch=coverstats::amaz...

I wonder how long that's going to last.